After the TMUA: How to Prepare for Your Oxbridge Interview

The TMUA is a Test of Mathematics for University Admission, and for Oxbridge candidates it is finished well before the decision it feeds. UAT-UK sends scores to universities “shortly after testing” and to candidates in November, about four weeks later. It sets no cut score. Cambridge then interviews between 7 and 18 December 2026, and Oxford its mathematics and computer science courses between 14 and 17 December 2026.

Your Score Reaches the Universities Before It Reaches You

The most useful thing to understand about the weeks after a TMUA sitting is that the two parties in this process are not working from the same information at the same time. UAT-UK, which runs the test jointly for Imperial College London and the University of Cambridge, describes the sequence in its own candidate handbook: “Test results will be sent to your UCAS choices shortly after testing. Results will be released to candidates in November”. Separately it states that “Scores are released around 4 weeks after your test date.”

Set against this cycle's dates, that produces a specific and slightly uncomfortable shape. The first sitting ran from Monday 12 October 2026 to Friday 16 October 2026. Universities had the scores within days. Candidates get them in November 2026, by which point shortlisting is under way or done. The quiet period between the test and an interview invitation is not an administrative lull; it is the part of the process that runs without you, on a number you have not seen.

Two practical consequences follow. The first is that there is nothing to chase: the score cannot be sent faster, withheld, or appealed into a different shape. UAT-UK notes that matching to your UCAS application is “done automatically so you do not need to do anything except making sure your personal details in UCAS and UAT-UK are the same.” The second is that the interview is the only remaining variable, and the calendar for it is already published.

This page is about that gap. It does not cover which course requires which admissions test — our guide to the ESAT, TMUA and TARA for 2027 entry sets out the mapping course by course, and the eight Oxbridge courses that take the TMUA are named there. What follows is about the stage that begins once the paper itself is behind you.

What Can Each University Actually See?

Candidates routinely assume more discretion exists here than does. UAT-UK is blunt on both halves of it: “You cannot hide your score from specific UCAS choices” and “You cannot choose which universities receive your score.” Registering for the test is consent to that distribution — “When you register for a test, you agree to your scores being shared.”

There is one genuine limit, and it is worth knowing precisely because it is often described wrongly. UAT-UK: “A university can only see your score if the course you applied for requires the UAT-UK test. This applies whether the test is optional or compulsory.” Scores are matched to courses rather than to institutions, so a fifth UCAS choice on a course that does not use the TMUA does not receive it. That is a fact about the matching mechanism, not a setting you can change.

What the receiving university does next is its own business, and both of ours say so plainly. Oxford describes the scale it is working at: it “typically receives over 23,000 applications for around 3,300 places every year and shortlists approximately 10,000 candidates”. Cambridge frames the interview stage from the other end: “We invite many applicants to interview. Not everyone interviewed will be offered a place, but everyone who is made an offer will have been interviewed.”

Read those two sentences together and the position after a TMUA sitting is clear enough. The test is one input into a decision about who is invited. Nobody receives an offer without the interview, so the interview is not a formality laid over a decision already taken on the paper.

Why There Is No TMUA Pass Mark

The single most persistent piece of misinformation about this test is that a threshold exists. UAT-UK addresses it directly: “It is not possible to fail UAT-UK tests. You should focus on doing your best. UAT-UK does not make any admissions decisions, or implement any cut score. It is entirely up to each institution how they use your test score but it is normally considered amongst lots of other aspects of your UCAS application.”

The reporting scale is published, which is more useful than a rumour. “The scale runs from 1.0 (low) to 9.0 (high) to one decimal place”, and UAT-UK anchors it: “The 50th percentile of the score distribution equates to 4.5 and the 90th percentile to 7.0.” Those are percentiles of the whole cohort sitting the test, not of any one university's applicants, and they move with the cohort.

Two mechanical details explain why a raw mark would not have meant anything anyway. Tests are delivered over a window, so “candidates sitting at different times and in different locations, see different versions of the test”, and “Scores are statistically equated using Rasch analysis” so that different versions can be reported on one scale. Your reported score therefore depends on the number of correct answers, the version you sat, and the performance of the cohort — which is why counting the questions you think you got right on the way out of the test centre predicts very little.

One difference from the ESAT is worth noting if you sat both. “For the ESAT, you will receive a separate score for each module. For the TMUA, you will receive a single overall score.” There is no per-paper breakdown to interpret and no weaker module to explain.

What the TMUA Could Not Measure, and What the Interview Measures Instead

This is the part worth spending the remaining weeks on, and it follows from the format rather than from anything mysterious about Oxbridge. The TMUA is multiple choice, timed, and silent. No one sees your working. No one tells you that you have taken a wrong turn. No one asks what you would do if a condition were relaxed. A score on it is a measure of accurate, fast, unassisted selection between given options.

Every one of those constraints is inverted at interview. Cambridge tells applicants that “You may be asked to share your workings if the interviewers ask you a problem-solving question. This usually happens if you've applied to a science or mathematics course.” The workings are the assessed object, not a means to an answer. Cambridge's Computer Science department is more explicit still about what happens when you stall: “Your interviewers will give you hints if you get stuck or go in the wrong direction”, and “It is absolutely fine to take some time to think about how to solve the problem before starting out and before talking about it.”

Oxford describes the same thing from the tutor's side. Its guidance says there “may well be more than one right way to answer a question in which case tutors will be more interested in exploring your thought process”, and that interviews let tutors see “how you think, respond to new ideas and engage with material beyond your school or college syllabus”. Cambridge lists what it is assessing, and the list is not a list of content: your understanding of the subject area, your readiness to study at a high academic level, whether you will thrive in the Cambridge learning environment, your ability to think critically and independently, your curiosity and openness to new ideas, and your enthusiasm for the subject.

The practical translation is narrow and unglamorous. Keep doing mathematics, but change three things about how: work out loud rather than on paper alone; work with someone who will interrupt and add a constraint mid-problem; and stop treating a hint as evidence that you have failed, because the published guidance treats responding well to a hint as part of what is being measured. A timed multiple-choice paper trains none of these, which is precisely why more of it is the wrong preparation now.

Oxford and Cambridge Run Different Interview Weeks in December 2026

A candidate holding applications at both universities needs the two calendars side by side, and they are not the same. Cambridge publishes a “Main interview period: 7 December to 18 December 2026” and, for those pooled, “Winter pool interviews: around mid to late January 2027”. Oxford's published timetable puts its mathematics and computer science courses on “Monday 14, Tuesday 15, Wednesday 16 and Thursday 17 December, although interviews for some colleges may start in the week commencing Monday 7 December.”

Two things in that table are commonly misread. Cambridge states that “The timing of your interview does not reflect your perceived strength as a candidate” and, separately, that “The number of interviews you have does not relate to how strong your application is.” A late slot and a short list of interviews are both routine variation by college and subject.

If you sat in the second window rather than the first, the calendar is different again: UAT-UK's Test Sitting 2 runs from Monday 4 January 2027 to Friday 8 January 2027, and results for that sitting are released to candidates in February. Most Oxford and Cambridge applicants are required to sit in October precisely so the score is available for December shortlisting; the January window exists for mature applicants to a Cambridge mature college with a January deadline, and for applicants to Oxford's Astrophoria Foundation Year.

Our subject pages carry the per-course detail. After the TMUA for Cambridge Computer Science covers the college-by-college formats and the extra written assessment two colleges set, and After the ESAT for Oxford Biomedical Sciences does the same for the ESAT side of the same fortnight.

Being Shortlisted Does Not Always Mean You Have Finished Sitting Things

This catches people out, because the natural assumption after a national admissions test is that the written stage is over. At Cambridge it sometimes is not. The university publishes a College admission assessments page for 2027 entry which states that “Some courses involve an admissions assessment that is taken after we shortlist for interview. If you are shortlisted for interview, you will take the relevant assessment. The College that interviews you will let you know when and how you should take your assessment.”

The administrative half is at least easy: “You do not need to register for College admission assessments”, and the interviewing college arranges everything. Some of these are sat in a named window before the interviews — the Architecture and Design assessment on Wednesday 18 November 2026, the Linguistics assessment on Tuesday 17 November 2026, the Modern and Medieval Languages assessment on Thursday 19 November 2026 — and others are set by individual colleges on the day of interview.

Which of these applies to you is a question about your course and your college, not about the TMUA, and the honest answer is that you have to look it up rather than infer it. Cambridge lists the courses it covers on that page and directs anyone whose course is not listed to the admissions assessments page instead. A candidate who assumes the October paper closed the written stage can discover otherwise in an invitation email in late November.

Oxford runs this differently: for the courses that take a UAT-UK test, the October paper is the written stage and the December interviews are conversations. Where a course requires submitted written work, that is asked for earlier in the cycle rather than added after shortlisting.